Major depressive symptoms in breast cancer patients with ovarian function suppression: a cross-sectional study comparing ovarian ablation and gonadotropin-releasing hormone agonists

We conducted an anonymous, cross-sectional study between July 2019 and June 2020 at seven hospitals across China. Eligible participants were female breast cancer patients who had either OA by bilateral oophorectomy or GnRHa by one of the following methods: monthly intravenous doses of 3.6 mg of goserelin acetate (Zoladex, AstraZeneca) or monthly intravenous doses of 3.75 mg of leuprorelin acetate (Leuplin, Takeda) or an intramuscular injection of 11.25 mg of leuprorelin acetate once every three months. Indications for ovarian function inhibition in breast cancer patients included premenopausal with either early or metastatic disease. In early breast cancer, endocrine therapy containing OFS is recommended for high-risk premenopausal hormone receptor-positive breast cancer, and should be considered for moderate risk patients. In metastatic hormone receptor-positive breast cancer, all the premenopausal patients should be converted to postmenopausal status and received endocrine therapy. Bilateral ovarian ablation and medical GnRHa had the similar effect and were superior to ovarian radiotherapy. Therefore, either bilateral ovarian ablation or medical GnRHa is indicated in premenopausal women with hormone receptor-positive breast cancer. The eligible participants were invited to complete the Breast Cancer Survivorship Ovarian Function Suppression Survey (OFS-Q5), an online questionnaire that included the Patient Health Questionnaire 9 (PHQ-9), which is an instrument for screening depressive symptoms, and the Female Sexual Function Index (FSFI) questionnaire. The validity and reliability of the Chinese versions of these questionnaires have been previously demonstrated [19, 45, 47]. The study was anonymized to improve the participation rate and accuracy of the responses.

Participants fitting the following criteria were included in our study: (1) a clear diagnosis of breast cancer (tissue or cell diagnosis), (2) age >18 years, and (3) recipient of OFS via surgical bilateral oophorectomy or GnRHa. The exclusion criteria were as follows: (1) refusal to provide informed consent, (2) inability to understand the questionnaires, and (3) undergoing ovarian irradiation. Written informed consent was obtained from each participant, and the study protocols were reviewed and approved by the ethics committee of the Cancer Hospital of China Medical University (protocol RB no. 20190545). The study was conducted in accordance with the principles of the Declaration of Helsinki.

The questionnaire was divided into four sections. The first section mostly comprised items for obtaining the respondents’ demographic and clinical characteristics, the second section recorded OFS information, and the following two sections respectively evaluated major depressive symptoms and sexual function in both cohorts. Two assistants were trained to develop a knowledge base and communication skills related to OFS and breast cancer. The first assistant explained and confirmed the information, step by step, with the patients, and the second assistant reconfirmed the responses via telephone 3–10 days after the survey’s implementation. The questionnaire was in Chinese, which was the respondents’ native language.

As previously noted ([16]), in PHQ-9 questionnaire, each of the nine items was assigned a score of 0–3, and the scores were then summed to obtain the final score. The following PHQ-9 score ranges have been recommended for determining levels of depression: 0–7: none and/or mild, 8–14: moderate, 15–19: moderate to severe, and 20–27: severe. Lower scores indicate better emotional functioning, whereas a sum score of ≥15 indicates major depression.

Sexual function was quantified using the FSFI questionnaire (Isidori AM et al 2010), which is a 19-item survey instrument that specifically assesses six sexual functioning domains (FSFI-1: desire, FSFI-2: arousal, FSFI-3: lubrication, FSFI-4: orgasm, FSFI-5: satisfaction, and FSFI-6: pain).

Data analysis was performed using the IBM SPSS Statistics for Windows, version 23.0 software package (IBM Corp., Armonk, N.Y., USA). Fisher’s exact test was used for categorical data. Differences in age and OFS duration were analyzed using unpaired t-tests. The means of the PHQ-9 total scores, FSFI total scores, and FSFI sub-scores were subjected to nonparametric Mann–Whitney U tests to determine whether the data were normally distributed. The multivariate analysis was performed with logistics analysis. P values < 0.05 indicated statistical significance.

As shown in Fig. 1, the final sample of patients with complete demographic characteristics and PHQ-9/FSFI scores comprised 563 individuals. Of these patients, 174 had undergone OA and 389 were being given GnRHa for OFS. Table 1 lists the main characteristics of the GnRHa and OA cohorts. The patients’ median ages in the OA and GnRHa cohorts were 46.0 and 43.0 years, respectively.

Table 1 shows that the time lapse from the commencement of ovarian suppression was significantly longer in the OA cohort than in the GnRHa cohort (30.3 ± 29.1 months vs. 22.8 ± 22.1 months, P = 0.003). The majority of respondents had low annual incomes (77.0%), had estrogen receptor or progesterone receptor positive breast cancers (97.2%), and were human epidermal growth factor receptor-2 (HER2)-negative (76.1%). In addition, 57.9% of the patients in the GnRHa cohort were educated up to or above college level, whereas the corresponding percentage for the OA cohort was only 40.8%. Patients receiving GnRHa favored breast-conserving surgery (26.0% vs. 16.7%, P = 0.016).

The results shown in Table 2 reveal that the mean PHQ-9 sum score of the GnRHa cohort was lower than that of the OA cohort (11.4 ± 5.7 vs. 12.8 ± 5.8, P = 0.079). According to the treatment algorithm for depressive symptoms, 36.2% (63/174), 28.2% (49/174), and 12.6% (22/174) of the patients in the OA cohort were respectively categorized in the moderate, major, and severe depressive symptoms groups. Notably, there were significantly fewer patients with major depressive symptoms (PHQ-9 ≥ 15) in the GnRHa cohort than in the OA cohort. A positive correlation was found between major depressive symptoms and OA, and the absolute difference was approximately 9.1 percentage points (31.1% vs. 40.2%, P = 0.025). Item-level responses indicated that 15 of the 174 patients who underwent OA and 30 of the 389 patients taking GnRHa experienced suicidal ideation, which is considered as a symptom of major depressive symptoms, but this difference was not statistically significant (8.6% vs. 7.7%, P = 0.713).

As shown in Table 3, patients receiving GnRHa had lower mean and median FSFI scores than patients who had undergone OA (mean: 17.8 ± 8.7 vs. 19.3 ± 8.5, P = 0.205; median: 17.8 vs. 19.6). The results for sexual dysfunction, which was defined as FSFI < 23, revealed that there was strong evidence of GnRHa-induced sexual dysfunction; 61.5% of OA patients met the criteria for sexual dysfunction compared with 72.2% of patients receiving GnRHa (P = 0.011). According to the data shown in Fig. 2, patients receiving GnRHa had slightly lower scores on most of the FSFI sub-scores compared with the scores of patients who had undergone OA. However, their scores of items in the pain section were significantly lower (FSFI-6 scores: mean ± SD: 3.3 ± 2.2 vs. 2.5 ± 2.2, P = 0.007), and there were also significant differences in scores of the item on lubrication (FSFI-3 scores: mean ± SD: 3.5 ± 2.5 vs. 2.9 ± 2.6, P = 0.048).

The data in Table 4 and 5 show that the type of ovarian suppression was closely associated with major depressive symptoms in the univariate and multivariate analyses (OA vs. GnRHa: Exp (B) = 1.805 [95% CI: 1.204–2.705], P = 0.004). As noted above, patients receiving GnRHa favored breast-conserving surgery (26.0% vs. 16.7%, P = 0.015). Mastectomy breast surgery was negatively correlated with major depressive symptoms in the multivariate analysis (Mastectomy vs breast-conserving: Exp (B)=0.461, 95% CI 0.301-0.707, P<0.001). As mentioned in Table 4, patients with fulvestrant combined with OFS had less major depressive symptoms in the univariate analysis. Nonetheless, the combined ET was not significantly correlated with major depressive symptoms (Exp(B)=0.720, 95% CI=0.379-1.366, P=0.314) in multivariate analysis in Table 5. Further analysis revealed that sexual dysfunction was negatively associated with major depressive symptoms (sexual dysfunction vs. normal: Exp (B) = 0.512 [95% CI: 0. 348–0.752], P = 0.001).

In the final part of the questionnaire, respondents were asked if they would choose to change to another kind of OFS. For example, “Would you choose OA instead of GnRHa or vice versa?” The majority of the respondents in both cohorts viewed this change positively. Approximately both more than half of the patients in both cohorts opted to change the type of OFS (GnRHa: 62.7% vs. OA: 60.8%). For concerns that costs could affect the final choice, we included the following item: “If cost was not a consideration, would you choose to change to another kind of OFS” In response to this question, the ratio of substitution showed a marked increase to 70.3% for the OA cohort, whereas the increase was not marked for the GnRHa cohort (67.1%). The majority of respondents in both cohorts may have been dissatisfied with their current OFS solution because of depressive symptoms, sexual dysfunction, or an overall decreased quality of life, and the majority of participants were reluctant to receive OA because of its cost.